Accepted author manuscript, 1.56 MB, PDF document
Research output: Contribution in Book/Report/Proceedings - With ISBN/ISSN › Conference contribution/Paper › peer-review
Research output: Contribution in Book/Report/Proceedings - With ISBN/ISSN › Conference contribution/Paper › peer-review
}
TY - GEN
T1 - A Two-Level Machine Learning Framework for Managing EV Charging and Renewable Energy Curtailment in Smart Grids
AU - Nasr Esfahani, Fatemeh
AU - Suri, Neeraj
AU - Ma, Xiandong
PY - 2025/4/16
Y1 - 2025/4/16
N2 - The increasing integration of electric vehicles (EVs) and renewable energy sources (RES) into power grids introduces significant challenges in managing dynamic energy demands and ensuring grid stability. This paper proposes a comprehensive two-level machine learning (ML) and optimisation framework for intelligent energy management in EV- and RES-integrated smart grids. In the prediction layer, supervised ML models, including Random Forest (RF) and Gradient Boosting (GB), accurately forecast EV charging demand and renewable generation. These forecasts are then fed into the optimisation layer, where a multi-objective particle swarm optimisation (PSO) algorithm minimises power losses, optimises EV charging schedules, and reduces renewable curtailment while ensuring voltage stability. The framework is evaluated on a modified IEEE 14-bus system incorporating EV charging stations, photovoltaics (PV), and wind turbines. Simulation results validate the effectiveness of the proposed framework, demonstrating a reduction in renewable energy curtailment and improved computational efficiency compared to benchmark optimisation methods.
AB - The increasing integration of electric vehicles (EVs) and renewable energy sources (RES) into power grids introduces significant challenges in managing dynamic energy demands and ensuring grid stability. This paper proposes a comprehensive two-level machine learning (ML) and optimisation framework for intelligent energy management in EV- and RES-integrated smart grids. In the prediction layer, supervised ML models, including Random Forest (RF) and Gradient Boosting (GB), accurately forecast EV charging demand and renewable generation. These forecasts are then fed into the optimisation layer, where a multi-objective particle swarm optimisation (PSO) algorithm minimises power losses, optimises EV charging schedules, and reduces renewable curtailment while ensuring voltage stability. The framework is evaluated on a modified IEEE 14-bus system incorporating EV charging stations, photovoltaics (PV), and wind turbines. Simulation results validate the effectiveness of the proposed framework, demonstrating a reduction in renewable energy curtailment and improved computational efficiency compared to benchmark optimisation methods.
M3 - Conference contribution/Paper
BT - ICCEP - 9th International Conference on CLEAN ELECTRICAL POWER
PB - IEEE
T2 - 9th International Conference on CLEAN ELECTRICAL POWER
Y2 - 24 June 2025 through 26 June 2025
ER -